tlipinski / bitwig-novation-impulse Goto Github PK

Hi,

Nice project going on. Is the functionality of your work discussed somewhere on a forum? Or on a youtube video?

Hello , i can't find the script file ?

I am trying to configure Moss's Midi Monitor, but when I try to select the Impulse as my chosen MIDI source, I get this error:

If I disable your device script, I can activate the Midi Monitor, but obviously I would like both to be working side by side. Interestingly, the keyboard is still sending MIDI, even with your script disabled, but presumably with a reduced feature set.

I asked Jürgen about this issue, and he suggested that your script may not be "multi-client".
Your thoughts? 🙂

The extension's initialization procedure sends a SysEx that establishes a template. The template creates several conflicts between assignable controls and non-assignable controls:

• In MIDI mode, Encoder 7 sends the same CC (0x1B) on the same channel (0) as the Rewind button.
• In MIDI mode, Encoder 8 sends the same CC (0x1C) on the same channel (0) as the Fast Forward button.
• In MIDI mode, Fader 1 sends the same CC (0x29) on the same channel (0) as Impulse's various "program change" controls (the + button, the - button, and the Data knob).

The extension interprets the Encoder 7 and 8 messages (in MIDI mode) as transport messages. As a result, the extension:

• Rewinds each time Encoder 7 sends the value 1.
• Fast Forwards each time Encoder 8 sends the value 1.

The extension currently doesn't handle MIDI Fader 1 or program change messages (CC 0x29), so it doesn't (yet!) behave weirdly for those.

I'm pondering several solutions for my own controller:

1. Assume that the CC message was sent by the non-assignable control.
   • This has the effect of making the faders and encoders act strangely in MIDI mode. The exact weirdness will depend on how the extension interprets values other than 0 and 1.
2. Assume that the CC message was sent by the assignable control.
   • The Rewind and Fast Forward buttons send only the values 1 (press) and 0 (release). The program change controls send only the value 0. If the plugin/device that ultimately responds to these MIDI messages expects values in the normal range (0–127), these limited, button-initiated values will be problematic.
3. Create extension Preferences to allow the user to choose the appropriate command for each conflicting CC message.
4. Initialize the Impulse with a pre-defined template that has no conflicting assignments.
   • I'm concerned that this may overwrite the template in the user's currently selected template slot.I haven't verified that this would happen.
5. Discover the encoders', faders', and buttons' MIDI-mode assignments by examining the current template.
   • Given that a user template may include assignments that conflict with each other and with non-assignable controls, I'll have to deal with the conflicts using solution 1, 2, or 3.

Another complication: What to do if the user switches templates while the extension is active. (This makes me lean toward solution 5, which alas is the most complex one).

Hi Tomasz
thanks so much for the scripting support. I am not very familiar with Java and I was wondering if you can do a new beta release with the clip launcher changes.

I have bravely convinced the controller to crash several times in a simple project with two FM8s and a Bitwig DrumMachine.

Each crash happened immediately after I switched focus from an instrument device (FM8) to a utility nested in the FM8's Post FX chain. One FM8 has a nested MeldaProduction MUtility. The other has a Bitwig Tool utility. I don't remember which of those nested devices I selected each time, so I don't know whether the crashes were related to a specific device.

At the moment, that's all I know about the variables involved. If I learn more, I'll let you know.

Does the Bitwig Controller API create a log when an extension crashes (or while it runs)? If I can find a log, I'll attach it.

Bitwig Studio 4.0.1
Novation Impulse 61
Windows 10

Using this controller script, sustain pedal events (cc64) on the Impulse 61 are not processed correctly.
Controller script console output:

[B2:64:127] -> unhandled
[B2:64:0] -> unhandled

Would be happy to submit a PR for a fix for this with a little bit of guidance. I assume there's just some sort of mapping missing?